Mirrors/ArchiveBox
2
0
Fork 0
mirror of https://github.com/ArchiveBox/ArchiveBox synced 2025-02-18 06:18:25 +00:00
Code Issues Projects Releases Packages Wiki Activity

catch json parse errors in link archiving

Browse source
This commit is contained in:
Nick Sweeting 2019-02-04 08:00:08 -08:00
parent c37941efd1
commit ad7038e031
1 changed files with 30 additions and 27 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

5
archivebox/archive_methods.py
View file

@ -80,6 +80,7 @@ def archive_links(archive_path, links, source=None, resume=None):
def archive_link(link_dir, link, overwrite=True): def archive_link(link_dir, link, overwrite=True):
"""download the DOM, PDF, and a screenshot into a folder named after the link's timestamp""" """download the DOM, PDF, and a screenshot into a folder named after the link's timestamp"""
try:
update_existing = os.path.exists(link_dir) update_existing = os.path.exists(link_dir)
if update_existing: if update_existing:
link = { link = {
@ -115,9 +116,11 @@ def archive_link(link_dir, link, overwrite=True):
if FETCH_MEDIA: if FETCH_MEDIA:
link = fetch_media(link_dir, link, overwrite=overwrite) link = fetch_media(link_dir, link, overwrite=overwrite)
write_link_index(link_dir, link) write_link_index(link_dir, link)
except Exception as err:
print(' ! Failed to archive link: {err.__class__.__name__}: {err}')
return link return link
def log_link_archive(link_dir, link, update_existing): def log_link_archive(link_dir, link, update_existing):